Re: Focus/fixation tolerance

[T3D Larry Berlin <lberlin@xxxxxxxxx>]

>Date: Thu, 26 Dec 1996 21:10:57 -0500 (EST)
>From: T3D  John O. Merritt  writes:
>
>.................. snip ...............
>It would be nice to have a family of curves showing how long
>it takes for various percentages of the population to become
>aware of discomfort when viewing various degrees of discrepancy
>between where the eyes are focused and where they are fixated.

************  A nebulous concept...but perhaps of general reading interest...

>
>This same cumulative plot could be used to present data on
>how long it takes for various degrees of vertical or rotational
>misalignment to create an awareness of discomfort.

************  This is far more fixed and measurable and meaningful...

>
>In addition to perceived discomfort, there wouild also be data
>on adverse after-effects such as headache, difficulty in normal
>stereo vision (fusing or focusing), diplopia, etc.

***********  I would like to suggest that any such survey generated
information will be necessarily fuzzy in importance because of the wide
range of physical ability and how those factors inter-relate with tiredness,
ambient light color and intensity, mental acuity of the moment, emotional
states, familiarity with the display parameters, time of day, etc. This is
one area that generalizations (especially scientifically devised surveys)
can be obtusely off mark. Primary in this potentially misleading information
is the term *perceived discomfort* which is an extremely flexible situation
and not at all fixed even in one individual.

The only obvious guideline is to arrange them as close to reality as
possible so as to minimize eye-strain. Since viewing stereo pairs as pairs
introduces abnormal eye positioning, this is the area where most likely
discomfort (for these factors) will be felt, but it's relative to one's
skill in the type of viewing in question. The average person off the street
who might be able to free-view but doesn't do it often will have a much
different reaction than someone who does it every day. The person off the
street, given lots of experience will report different parameters which
still will not be the same for every day tested.

Many of those who free-view often and for long periods claim no discomfort
or after-effect (a common situation with obvious discrepancy between focus
and fixation). I on occasion do experience either discomfort or temporary
difficulty in normal vision after extended stereo viewing. I solved my
problem by doing both crossed and parallel viewing regularly for improved
muscle tone. This indicates that basing *rules of thumb* on levels of
*perceived discomfort* is fundamentally the wrong benchmark. With the new
interlaced viewing systems, these effects are minimized completely and the
range of depth available should be correspondingly greater. This changes the
picture to consideration of factors like the weight and stability of the
specific shutter glasses being used. It gives a decided preference for
passive systems in which far simpler glasses can be used. For anaglyph
systems, again it is external factors like the red/blue rivalry that
introduces the most strain.

A perceived limit can often be overcome by better arranging the content of
the images, the framing of the scene, or some other adjustment, negating the
idea that a certain situation (narrowly defined) is supposedly not a good
one to use. Since viewing stereo pairs can be done relatively easily and for
extended periods of time, the idea of a mismatch between focus and fixation
in images coincident to the screen is almost beyond possibility, other than
abnormal extremes that would be mismatched in most methods of viewing.

>
>I will try to nail down this issue of focus/fixation mismatch
>by reviewing any suggested references, and making up a table 
>form of annotated bibliography.  This would include rules of
>thumb and anecdotal information as well as carefully controlled
>experimental data.

***********  focus/fixation mismatch is a term that is hard to define
because as long as the focus range is available to the viewer, and the
fixation depths of image elements aren't totally out of range there is no
*mismatch*. Perception centers on the the area in which the information
coincides. Other factors can mismatch the information but that's not a valid
issue to focus/fixation.

>
>The design of the stereo display parameters in video systems
>used for relatively long continuous periods of time, such as
>undersea teleoperators, telesurgery, hazardous materials handling,
>remote inspection, etc., depends on solid answers to this focus/
>fixation mismatch issue.          

***********  If you are talking about paired image viewing-by-eye technique
(x'd or parallel), the question is valid, but for interlaced images or any
system that has the center of the image depth at the screen or close to it,
it's a situation with a much greater overall latitude. It becomes an issue
of image composition and arrangement rather than focus/fixation. The
operations you mention here are perhaps less stressful overall when done in
stereo as opposed to 2D, relating to the more available information rather
than focus/fixation issues.

>
>Too little caution will result  in rejected and cursed 3D display 
>systems, whereas too much caution results in very shallow depth
>viewing volumes, or overly conservative camera separation (with
>resulting limited depth acuity).

**********  Again, because interlaced viewing centers on the screen instead
of behind or in front, as in paired viewing, the issue of focus/fixation has
far less impact. The use of camera separation for greater depth acuity is
something that should be done especially where accurate perception of depth
relationships outweighs the consideration of *Ortho-ness*. The resulting
views will not usually challenge the relationship of focus/fixation for
interlaced or similar images which are essentially coincident to the screen
as the central depth for video/monitors.
___________________________________
>From: T3D john bercovitz  <bercov@xxxxxxxxxxx>
>To: tech3d@xxxxxxxxxxxxxxx
>Subject: Re: Focus/fixation tolerance
>Message-ID: <199612270316.TAA24918@xxxxxxxxxxx>
>
>I think it would be interesting to add acquisition or fusion
>time to the cataloged parameters.  I would think degrees of mismatches 
>would cause degrees of delay which could have a maximum acceptable 
>limit when the usage is control of an RPV, for instance.
>

**********  This would be interesting additional information but only
applicable if the display system in question uses side by side pairs of one
sort or another. The delay experienced by a viewer has more to do with
patterns of expectation and familiarity than with a fixed degree of
difficulty. Once an operator is into a particular set of viewing parameters,
acquisition of the next frame of information relates to changes from the
previous situation not to factors of absolute f/f. Again, if the system uses
images that are coincident with the screen depth, re-acquisition time for an
operator who looks away will be almost instant, especially if the operator
was familiar with the viewing parameters when last seen.

3D perception is a very flexible and relative phenomenon. With interlaced
viewing systems, 3D is ready to be used and promises to be as easy to view
as 2D, and easier where depth information is valuable to the working process.

Larry Berlin

Email: lberlin@xxxxxxxxx
http://www.sonic.net/~lberlin/
http://3dzine.simplenet.com/


------------------------------

End of TECH3D Digest 70
***********************
***********************
 Trouble? Send e-mail to 
 wier@xxxxxxxxxxxxxxx 
 To unsubscribe select one of the following,
 place it in the BODY of a message and send it to:
 listserv@xxxxxxxxxxxxxxx 
   unsubscribe photo-3d
   unsubscribe mc68hc11
   unsubscribe overland-trails
   unsubscribe icom
 ***********************